提出了状态空间双线性系统的极大似然辨识方法。得到了以输入-输出序列为条件概率的似然函数解析表达式,推导了极大化似然函数的参数矩阵计算公式,给出适用于双线性系统状态估计的改进卡尔曼滤波方法,以及辨识系统参数的迭代估计算法。最后进行了数值仿真,结果说明了该方法的有效性。
Maximum likelihood identification is proposed for parameter estimation of state-space bilinear systems.The likelihood function conditioned on input-output series is constructed.Moreover,the parameter matrix is determined by the maximization of the likelihood function,and the modified Kalman filter suitable for state estimation of bilinear systems is presented.In addition,iterative parameter estimation algorithm for maximization of likelihood function is also given.Finally,numerical simulation is implemented and the results show the effectiveness of the proposed method.